Fast Mining and Updating Frequent Itemsets
Chaohui Liu, Jiancheng An · 2008
In order to overcome the drawbacks of apriori algorithm for mining frequent itemsets, TIMV (Three-dimensional Itemsets Matrix and Vectors) algorithm was proposed, which used three -dimensional itemsets matrix and vectors, and broke through the bottom-up framework of Apriori. Only needed one pass to scan the database and did not create candidate itemsets, we could gain all the frequent itemsets. Furthermore, this paper introduced FUFIA (fast updating frequent itemsets algorithm), which could get the new frequent itemsets through searching three-dimensional itemsets matrix when the database and the minimum support were changed. Both theoretical analysis and experimental results showed the feasibility and effectiveness of the two algorithms.